Quick Answer
For most fixed indoor and outdoor surveillance, a camera with a 90°–110° horizontal field of view covers a room corner or building face without wasting pixels. Go narrower (30°–60°) when you need to identify faces or plates at distance, and wider (120°–180°) only for situational awareness, because every degree of width costs pixel density. When one camera must do both jobs, a varifocal model lets you set the view at install time.
Field of view is the first spec that decides whether a camera actually does its job — before resolution, before low-light performance, before analytics. It determines what the sensor sees and, just as importantly, how many pixels land on each meter of scene. This guide is for integrators and facilities teams choosing fixed cameras for commercial sites; it covers how FoV interacts with resolution, when wide is a mistake, and how to read manufacturer FoV specs. If you are sizing recording alongside coverage, the retention and storage calculator pairs with this exercise.
What Field of View Actually Measures
Horizontal FoV is the angular width of the scene the lens projects onto the sensor, quoted in degrees. Manufacturers usually publish it as H / V / D — horizontal, vertical, and diagonal. The horizontal number is the one that matters for coverage planning; the vertical figure mostly determines how much floor and ceiling you capture.
The math is simple: the width a camera covers equals 2 × distance × tan(FoV÷2). A 100° camera looking 10 meters out sees a scene roughly 24 meters wide. A 34° view at the same distance sees about 6 meters. Same sensor, four times the pixel density on target.
Bold rule of thumb: doubling the angle roughly halves your usable identification range. Wide and sharp is a trade, not a free lunch.
Pixel Density: Why Wider Is Not Automatically Better
Video quality on target is measured in pixels per meter (px/m). Industry DORI guidance runs approximately: 25 px/m to detect that something is present, 63 px/m to observe behavior, 125 px/m to recognize a known person, and 250 px/m to identify an unknown one.
Work an example. A 4MP camera has about 2,560 horizontal pixels. Set to a 110° view, at 10 meters it spreads those pixels across a 28.6-meter scene — about 90 px/m: good observation, marginal recognition. The same camera at a 40° view concentrates them across 7.3 meters — about 350 px/m: identification grade. Neither setting is wrong; they are different jobs.
This is why a parking lot design usually pairs a wide camera for scene awareness with a narrow one on the entrance where plates and faces pass a choke point.
Matching FoV to the Job
Standard coverage, 80°–110°: the workhorse range for lobbies, docks, corridors seen from a corner, and building faces. The Hanwha QNV-8020R 5MP vandal dome sits here with an 80° horizontal view — a 5MP sensor at 80° holds recognition-grade density past 15 meters. Shop the IP camera range →
Wide fixed, 110°–130°: for open rooms and intersections where one camera must sweep a broad area. The Hanwha XNV-C6083R covers 120° horizontally; its 2MP sensor is honest about the trade — treat it as an awareness camera, not an identification tool beyond short range. The Hanwha ANO-L7012R 4MP bullet takes the same idea outdoors at 99° with more resolution behind it.
Panoramic, 180°: a single-sensor panoramic like the Vivotek CC831-HV panoramic dome watches an entire wall-mounted hemisphere — ideal for a retail floor or cafeteria where the question is “what happened where,” never “who exactly was that at 30 meters.”
Varifocal, set at install: when the mounting position is fixed before the coverage question is settled, a varifocal camera is insurance. The Axis P3277-LVE 5MP outdoor dome adjusts from 104° down to 34° — one SKU that can be a wide lobby camera or a narrow gate camera depending on how the installer leaves it.
Mounting Height Changes the Math
FoV planning on paper assumes the camera looks straight down the scene. Real cameras mount 3–5 meters up and tilt down, and that tilt does two things to your coverage plan.
It creates a dead zone under the camera. The steeper the tilt, the closer the blind cone under the mount. A camera at 4 meters tilted to watch a door 8 meters away typically cannot see the 2–3 meters directly beneath itself — which is exactly where someone stands to tamper with it. Cover each camera’s dead zone with its neighbor, not with hope.
It burns vertical FoV on floor and sky. A camera with 55° vertical coverage tilted downward spends a slice of it on the ground in front of the mount. This is why the vertical spec matters at high mounts: run out of vertical angle and faces at the far end of the scene crop out of frame just when they get close enough to identify.
Height also flattens faces. Above roughly 5 meters, the camera looks down at the top of heads instead of at faces, and no amount of horizontal FoV or resolution recovers an angle the geometry never captured. For identification points, mount lower and accept the vandal-resistant housing trade — that is what IK-rated domes are for.
Corridor Format: The Overlooked FoV Trick
Most sensors are wider than they are tall — 16:9. Hallways, aisles, staircases, and perimeter fence lines are the opposite: tall and narrow, or long and thin. Pointing a landscape sensor down a warehouse aisle wastes most of its pixels on the racking either side.
Many commercial cameras — including the Axis and Hanwha models above — support corridor format: the image rotates 90° so the long axis of the sensor runs along the long axis of the scene, turning a 16:9 stream into 9:16. The same sensor that covered 9 meters of aisle width it did not need now covers those pixels down the aisle’s length. For corridors, aisles, and platform edges, enabling corridor format is frequently worth more than a resolution upgrade — it is free pixel density in the direction that matters.
Common FoV Mistakes on Site Plans
Buying the widest lens “to be safe.” The most common error in DIY-drawn plans. Wide angles feel like insurance, but the premium is paid in pixel density — the footage exists and proves nothing. Design from the evidence requirement backward.
Counting the diagonal spec as coverage. A “114°” camera that is 114° diagonal may be 95° horizontal. Two of those on a building face, planned at the diagonal figure, leave a seam neither camera covers.
Ignoring overlap. Adjacent coverage cameras should overlap 10–15% so a subject cannot walk a seam between views, and so each camera watches its neighbor’s dead zone and mount.
One camera, two jobs. A single camera asked to both watch a lot and identify at its entrance does neither. The math above is unforgiving: the settings that make one job work break the other. Split the roles; the second camera is cheaper than the incident nobody can prosecute.
Every Pick, Side by Side
| Camera | Horizontal FoV | Resolution | Best For |
|---|---|---|---|
| Axis P3277-LVE | 104°–34° varifocal | 5MP | One SKU, coverage decided at install |
| Hanwha QNV-8020R | 80° | 5MP | Standard entrances and corridors |
| Hanwha ANO-L7012R | 99° | 4MP | Wide outdoor faces, low light |
| Hanwha XNV-C6083R | 120° | 2MP | Broad awareness, short-range detail |
| Vivotek CC831-HV | 180° panoramic | 5MP | Whole-room retail and common areas |
Reading Manufacturer FoV Specs Without Getting Burned
Check which number is quoted. A spec sheet leading with the diagonal figure reads 15–25% more impressive than the horizontal one. Coverage plans are built on horizontal.
Varifocal specs quote a range — 104°–34° means wide end to tele end. The wide-end number is the marketing headline; confirm the tele end covers your identification distance.
Panoramic pixel density is not linear. A 180° single-sensor camera spends pixels unevenly across the scene; density at the edges falls off. Plan identification zones near the center or with a dedicated narrow camera.
Wide lenses distort. Below roughly 2.8mm focal length expect barrel distortion at the edges; dewarping helps live operators but recorded evidence keeps the raw geometry in mind.
Deployment takeaway: design from the identification points backward. Put narrow or varifocal cameras on the choke points where you must know who — doors, gates, lanes — then fill the spaces between with 90°–110° coverage cameras and let panoramics own the open rooms. A site plan that starts with “how few cameras can see everything” ends with footage of everything and evidence of nothing.
One more planning habit worth adopting: walk the site with the spec sheet numbers in hand. Stand at each proposed mount, look toward the farthest point you care about, and ask what job that view must do — detect, observe, recognize, or identify. The DORI density each answer demands, divided into the camera’s horizontal pixel count, tells you the widest angle you can afford at that distance. Ten minutes of this on-site arithmetic catches the mistakes that show up otherwise as unusable footage after the first incident.
Decision Shortcut
If you know the mounting point and the target distance, pick the FoV that puts at least 125 px/m on the target at that distance. If the mounting point is fixed but the job might change, take the varifocal. If the job is “see the whole room,” take the panoramic and accept that identification happens somewhere else.
FAQ
Does higher resolution fix a too-wide field of view?
Partly. Doubling megapixels claws back some density, but a 12MP sensor behind a 180° lens still spreads thin at distance — and drives storage costs up sharply. Match the angle to the job first, then buy resolution. The storage calculator shows the retention cost of each step up.
What FoV do I need to read license plates?
Plate capture is a narrow-angle job: typically 10°–35° aimed at a lane choke point, with shutter and IR settings mattering as much as the angle. A general 100° coverage camera over a parking lot will not produce readable plates at distance.
Is one 180° camera cheaper than two 90° cameras?
On hardware and cabling, usually yes — and for awareness coverage that is often the right call. You give up pixel density and a second viewing angle, which matters when an incident needs a face or a plate rather than a timeline.
Does infrared range depend on field of view?
Yes — the same IR emitters spread across a wider angle illuminate a shorter distance. A spec sheet’s IR range is quoted at a specific lens setting; a varifocal camera at its wide end lights less distance than the tele figure implies. Outdoor wide-angle night coverage often needs supplemental IR.
Can I change field of view after installation?
Only on varifocal and PTZ models. Fixed-lens cameras are committed at purchase, which is why the mounting plan should exist before the order does.
Updated August 2026